Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Apr 25.
Published in final edited form as: Chem Biol Interact. 2020 Mar 3;321:109025. doi: 10.1016/j.cbi.2020.109025

Triterpenoid corosolic acid modulates global CpG methylation and transcriptome of tumor promotor TPA induced mouse epidermal JB6 P+ cells

Rasika R Hudlikar a,b, Davit Sargsyan a,b, Renyi Wu a,b, Shan Su b, Meinizi Zheng a,b, Ah-Ng Kong a,b,*,*
PMCID: PMC7238714  NIHMSID: NIHMS1572695  PMID: 32135139

Abstract

Epigenetic regulation is one of the driving forces in the process of carcinogenesis. Corosolic acid (CA); triterpenoid abundantly found in Lagerstroemia speciosa L. is known to modulate various cellular process including cellular oxidative stress and signaling kinases in various diseases, including skin cancer. Genetic mutations in early stages of skin cancer are well-documented, the epigenetic alterations remain elusive. In the present study, we identified the transcriptomic gene expression changes with RNAseq and genome-wide DNA CpG methylation changes with DNA methylseq to profile the early stage transcriptomic and epigenomic changes using tumor promoter TPA-mediated mouse epidermal epithelial JB6 P+ cells. JB6 P+ cells were treated with TPA and Corosolic acid by 7.5uM optimized by MTS assay. Differentiated expressed genes (DEGs) and Differentially methylated genes (DMRs) were analyzed by R software. Ingenuity Pathway Analysis (IPA) was employed to understand the differential regulation of specific pathways. Novel TPA induced differentially overexpressed genes like tumor promoter Prl2c2, small prolin rich protein (Sprr2h) was reported which was downregulated by corosolic acid treatment. Several cancer related pathways were identified by Ingenuity Pathways Analysis (IPA) including p53, Erk, TGF beta signaling pathways. Moreover, differentially methylated regions (DMRs) in genes like Dusp22 (Dual specificity protein phosphatase 22), Rassf (tumor suppressor gene family, Ras association domain family) in JB6 P+ cells were uncovered which are altered by TPA and are reversed by CA treatment. Interestingly, genes like CDK1 (Cyclin-dependent kinases 1) and RASSF2 (Ras association domain family member 2) observed to be differentially methylated and expressed which was further modulated by corosolic acid treatment, validated by qPCR. Given study indicated gene expression changes to DNA CpG methylation epigenomic changes modulated various molecular pathways in TPA-induced JB6 cells and revealed that CA can potentially reverse these changes which deciphering novel molecular targets for future prevention of early stages of skin cancer studies in human.

Keywords: Early stage skin cancer, Next generation sequencing (NGS), RNAseq and DNA methylseq, Ingenuity pathway analysis, Corosolic acid, TPA

1. Introduction

Melanoma and non-melanoma skin cancer (NMSC) are the most common types of cancer in the USA populations and known to occur with highest incidence [1]. Therefore, studying and understanding the mechanism of skin cancer promotion events and progression is considered crucial in order to elucidate the early molecular markers. Various epigenomic regulation including methylation, noncoding RNAs, and histone modifications, have been reported in promotion and progression of skin cancer [2–4]. Use of in vitro and in vivo models to study the promotion events and to screen potential preventive agents is important [5,6]. One well established in vitro system of JB6 murine epidermal model representing preneoplastic to neoplastic progression, which consists of promotion-resistant (P-) and promotion-sensitive (P +) preneoplastic cell lines as well as irreversibly transformed, tumorigenic cell lines [7,8]. Tumor promoter 12-O-tetradecanoyl phorbol-13-acetate (TPA) is known to induce the oxidative stress in skin cells to induce the clonal expansion of the initiated cells to induce the process of promotion. Transformation of preneoplastic epidermal JB6 cells with TPA is well established in vitro model to understand the process of tumor promotion and progression [9,10].

Epigenetic deregulation has been increasingly recognized as a hallmark of cancer during the last decade [11,12]. Various naturally occurring phytochemicals have been shown to modulate the process of cancer chemoprevention by altering the diverse epigenetic process including DNA CpG methylation [13–15]. A triterpenoid, Corosolic acid (CA) (Fig. 1A), also known as 2α-hydroxyursolic acid, is found Schi-sandra chinensis, Eriobotrya japonica, Lagerstroemia speciose L., Orthosiphon stamineus, and Weigela subsessilis [16]. CA has been reported to modulate the process of carcinogenesis in various cancers including colon [17], cervical cancer [18] and glioblastoma [19]. CA induces apoptosis in colorectal cancer cells HCT116 through caspase-dependent pathway [16]. It also enhances the antitumor effects of chemotherapy on ovarian cancer by inhibiting STAT-3 signaling [20]. CA have been shown to inhibit the proliferation of glioblastoma cells, U373 and T98G and activate the STAT3 and NF-κB in both human macrophages and glioblastoma cells [19]. CA also known to enhance the activity of 5-fluorouracil via m-TOR inhibition in SNU-620 human gastric carcinoma cells [21] and inhibits hepatocellular carcinoma cellular migration by modulating VEGFR2/Src/FAK pathway [22]. Our earlier study showed that the anchorage-independent growth of prostate cancer TRAMP-C1 cells was blocked by CA potentially via induction of mRNA and protein expression of Nrf2, heme oxygenase-1 (HO-1) and NAD(P)H Quinone Oxidoreductase 1 (NQO1). We further showed that CA decreased the level first five CpG sites of the Nrf2 gene promoter by DNA methylation, potentially via modulating the expression of DNMTs and HDACs [23]. However, despite of numerous studies to date have focused on understanding the anticancer activity of CA, the mechanism of this compound on the TPA-induced promotion of skin carcinogenesis remains elusive. In this study we analyzed the transcriptomic and CpG methylomic changes and preventive effects associated with CA treatment using JB6 P+ cells. This study will potentially identify the gene expression and methylation biomarkers in the process of skin carcinogenesis and chemoprevention using dietary phytochemicals which will provide promising prevention strategies for the future clinical trials.

Fig. 1.

Fig. 1.

Open in a new tab

Cell viability assay of JB6P + cells when treated with Corosolic acid. A) Chemical structure of Corosolic acid B) JB6 P+ cells were treated with different concentrations of CA for 1, 3 and 5 days as described in the Methods. Cell viability was determined by MTS assay. The data are presented as the mean ± SD of five observations.

2. Material and methods

2.1. Materials

Fetal bovine serum (FBS), Minimum essential medium (MEM), Penicillin/streptomycin antibiotic, and trypsin-EDTA were by Gibco lab (USA). Corosolic acid (CA) and dimethyl sulfoxide (DMSO) were obtained from Sigma-Aldrich (USA). TPA was procured from Alexis Biochemicals (USA).

2.2. Cell culture and treatment

The JB6 P+, mouse epidermal cells, was obtained from the American Type Culture Collection (ATCC; USA). MEM with 5% FBS was used for cell culture with humidified 5% CO2 and grown to 80% con-fluence for all the assays. Treatment with Corosolic acid (2.5–15 μM) and/or TPA (10 ng/ml) was done for 5 days. Cells treated with 0.1% DMSO was used as control.

2.3. Cell viability (MTS) assay

96-wells plate was seeded by JB6 P+ with 1 × 103 cells/well density for 24 h and were treated with DMSO or Corosolic acid at concentrations ranging from 2.5 to 15μM for 1, 3 and 5 days, replacing the medium every other day. The cell viability was determined using CellTiter 96® AQueous One Solution Assay (USA) as previously described [24].

2.4. RNA-seq sample preparation and data analysis

The JB6 P+ cells were treated with TPA or TPA + CA for 5 days and total RNA was isolated from cells using Allprep DNA/RNA Mini Kit. The quality of extracted RNA was determined using Agilent 2100 Bioanalyzer. The RNA library was constructed using An Illumina TruSeq RNA preparation kit (USA) was used for library preparation. The sequencing was performed using Illumina NextSeq 500 with 75 bp single end reads. The samples generated 30–40 million reads per sample. Adapter sequences were removed using Cutadapt [25]. Duplicate PCR sequences were removed and the mouse genome (mm10) alignment with HISAT2 was done as mentioned [26]. FeatureCounts (version 1.5.1) was used for counting genomic features with overlapping reads [27] and they were further analyzed for differential expression using DEGSeq (version 1.36.0) in R (version 3.4.0) [28]. Differential expression with log2 fold change was considered significant if p values were less than 0.05.

2.5. DNA methyl-seq sample preparation and data analysis

The JB6 P+ cells were treated with TPA or TPA + CA for 5 days and total DNA isolation was done using Allprep DNA/RNA Mini Kit. The methyl-seq was carried out as described previously [29]. Briefly, bisulfite conversion of extracted DNA was performed using an EZ DNA Methylation-Gold Kit (Zymo Research, USA) as described in the manufacturer’s protocol. Further enrichment of these DNA samples were done using an Agilent Mouse SureSelect Methyl-seq Target Enrichment System (USA). Illumina NextSeq 500 was used for DNA sequencing with 76-bp single-end reads, generating 30–40 million reads/sample. The DNA reads alignment was matched with the in silico bisulfite-converted mouse genome (mm10) with the Bismark alignment algorithm(version0.15.0) as described [30]. Following alignment, methylation counts were extracted using DMRfinder (version 0.1) and clustered the CpG sites into differentially methylated regions (DMRs) [31]. Each DMR contained at least three CpG sites. Differences in the methylation greater than 0.10 with p values less than 0.05 were considered significant. ChIPseeker (version 1.10.3) was used for genomic annotation with in R interface (version 3.4.0) [32].

2.6. Ingenuity Pathway Analysis (IPA)

Genomic isoforms with log2 fold changes > 1 or < −1 and q value < 0.05 were subjected to IPA (Qiagen). The input isoform mapping to the IPA knowledge base was done to understand the biological functions, networks, and pathways related to TPA and CA treatments.

2.7. Validation by quantitative PCR

The mRNAs significantly modulated by TPA and TPA + CA treatment were determined using quantitative PCR (qPCR) analysis. RNA (previously used for RNA-seq library preparation) was reversed-transcribed by TaqMan and analyzed by the QuantStudio 5 Real-time PCR system using SYBR Green PCR Master Mix (Thermo Fisher Scientific, Waltham, MA). The relative fold change was normalized to the internal standard gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH).

2.8. Statistical analysis

The MTS and qPCR data are presented as the mean ± standard deviation with observations in triplicates. P value less than 0.05, and false discovery rate (FDR) less than 10% were considered significant, except as otherwise indicated.

3. Results

3.1. Effect of CA on cytotoxicity of JB6P + cells

The MTS cell viability assay was performed for JB6 cells, as shown in Fig. 1B. Various concentrations of CA were assessed using JB6 P+ cells for 1, 3 and 5 days to determine the dose and time dependent effects on the cell cytotoxicity. The cell viability of JB6 P+ cells was less than 50% with CA (> 7.5uM) treatment post 3 and 5 days. Hence, 7.5 μM CA with 5 days treatment was used for the subsequent assays.

3.2. Effect on changes in transcriptomic profiling during TPA induced transformation in JB6 P+ cells treated using CA

In order to understand the changes in the gene expression profiling in TPA induced JB6 P+ cell transformation and effects of CA, RNA-seq was performed from cells treated with DMSO, TPA and TPA + CA groups. The gene expression data in all the three groups showed similar bimodal skewed histogram presenting the homogenous representation of the numerical counts in all three groups (Fig. 2A). The boxplot in Fig. 2B demonstrated the similar raw count distribution with upper and lower quartiles between the analyzed samples. To understand the clustering between the samples, Euclidean distance clustering analysis was performed. Fig. 2C indicated the apparent sample to sample separation (DMSO vs TPA vs TPA + CA). The orthogonal transformation was performed by plotting principle component analysis (PCA), showed marked separation of control, TPA and TPA + CA groups (Fig. 2D) signifying the strong impact of CA treatment on TPA induced JB6 P+ transformation. Further, the comparison of gene expression profiles was done between the groups (Control vs TPA and TPA vs TPA + CA) with the q value cutoff of less than or equal to 0.05 with log2(fold change) with threshold > 1.0 or < −1.0. The MA plot depicted the comparison of the read counts of both upregulated and downregulated genes in the Control vs TPA groups (Fig. 2E) and TPA vs TPA + CA (Fig. 2F). Fig. 2E represented higher number of upregulated differentially expressed genes (DEGs) depicting the strong impact of TPA treatment on JB6 P+ cells as compared to the control group. Interestingly, as represented in Fig. 2F, the comparison of TPA vs TPA + CA showed higher number of downregulated DEGs depicting the contrasting effect of CA treatment on TPA transformed JB6 P+ cells. These results are suggestive evidence that TPA alters the gene expression in mouse epidermal cells by upregulation of DEGs while CA downregulates the DEGs, however the impact of TPA treatment is stronger than CA.

Fig. 2.

Fig. 2.

Fig. 2.

Open in a new tab

Transcriptomic gene expression RNA-seq analysis of JB6P + cells treated by TPA and CA in JB6 cells. A) DEGs distribution by number of genes represented by count histogram. B) Boxplot showing between the sample distribution of annotated data (normalized) for the control, TPA, and TPA + CA groups. C) Pearson correlation amongst the groups clustered by Euclidean distance. D) Principal component analysis of RNA expression between the expression profiles of the control group are separate from those of the TPA and TPA with CA groups. E) MA plot for differential expression analysis between control vs TPA F) MA plot for differential expression analysis between TPA vs TPA + CA with threshold of q < 0.05 and log2(fold change) > 1 or < −1.

3.3. Effect on overall DEGs by TPA and its modulation by CA treatment

The effect of CA treatment on TPA induced neoplastic changes, the gene expression profiles were compared from TPA vs control group and the TPA + CA versus TPA alone group. The cutoff of q value < 0.05 and log2 fold change > 1.0 or < −1.0 were used to identify the DEGs. We identified deregulated 511 DEGs between control and TPA treated group (382 genes were upregulated while 129 genes were downregulated by TPA). We further identified the 152 DEGs deregulated between TPA + CA group and TPA only group (30 genes upregulated by CA while 122 genes were downregulated by CA) (Fig. 3A). Among the genes regulated by TPA and CA treatment, 89 genes were significantly reversed the gene expression (Fig. 3A). Top 52 genes differentially expressed by TPA and reversed by CA treatment are plotted in the heatmap (Fig. 3B), while the 89 genes which are differentially expressed by TPA and reversed by CA. Prolactin-2C2 (Prl2c2) growth and angiogenesis factor, Sprr2h (small proline-rich protein 2) known to be regulated by IL6/STAT3 signaling, Mrc1 (Macrophage mannose receptor1), Dkk2 (Dickkopf-related protein 2) which is WNT signaling pathway inhibitor, NOS2 (Nitric oxide synthase 2), CYP27B1 (cyto-chrome P450 family 27 subfamily B member1) are some of the major genes found to be differentially modulated by CA treatment as compared to TPA only. These overlapping DEGs have molecular importance in the regulation by CA as chemoprevention mechanism on TPA induced transformation on JB6P + cells.

Fig. 3.

Fig. 3.

Open in a new tab

Overview of differentially expressed genes (DEGs) between TPA and CA. A) Venn diagram of comparing the overexpressed and downregulated genes in both the control versus TPA group and the TPA only vs. TPA + CA groups. Genes with q < 0.05 and threshold of log2 (fold change) > 1 or < −1 were considered for the given analysis. B) Heat map showing the DEGs in both control vs TPA and TPA vs TPA + CA comparisons.

3.4. Effect on canonical pathways expression by Ingenuity Pathway Analysis (IPA) on TPA induced neoplastic transformation in JB6 P+ cells treated using CA

To understand the modulation of biological pathways associated with the DEGs, the canonical pathway analysis was conducted using IPA. The list of differentially regulated genes identified by a false discovery rate (q) < 0.05 and a log2 fold change cutoff of > 1.0 or < −1.0 to analyze 2034 the genes from the TPA group versus control group comparison. While threshold of p < 0.05 and log2 fold change > 0.1 or < −0.1 to analyze 2224 genes comparison amongst TPA + CA group versus TPA group. After obtaining the lists of differentially modulated pathways from these two comparisons (Supplementary Table 1), based on −log (P-value) and their ‘activation Z score’, top 20 common pathways were plotted using heatmap as shown in Fig. 4. Among these pathways, 13 pathways have shown to be upregulated by TPA and which was inhibited by CA treatment while 7 pathways were downregulated by TPA and activated by CA. To note, the modulated pathways were majorly of cell cycle regulatory, cancer and cell differentiation, apoptosis, signaling kinases associated pathways including p53 signaling, Erk signaling, TGF beta signaling etc. Interestingly, rho family GTPase, PEDF signaling pathway, TNFR2 signaling pathway which was upregulated by TPA treatment was further downregulated by CA treatment. Thus, the observed modulatory effects on CA on TPA induced transformation is attributed to inflammation, cell differentiation, apoptosis.

Fig. 4.

Fig. 4.

Open in a new tab

Heatmap demonstrating the top differentially regulated pathways in control vs TPA and TPA vs TPA + CA comparisons evaluated using Integrated pathways analysis (IPA).

3.5. Effect of CA on global DNA methylation on TPA induced transformation in JB6P + cells

Global alterations in DNA methylation were identified by CA treatment in TPA induced transformation in JB6 P+ cells methyl seq by implementing single base-pair resolution. The distribution of DMRs, as shown in Fig. 5A, are found to be in the promoter (< 1 kb), intron region and distal intergenic region (> 3 kb upstream of the transcription start site or downstream of the 5′ untranslated region (UTR). Similarly, the distribution of number and region of CpGs, the number of CpGs was greater in the promoter region than that in other regions such as in the gene body, 3 ′UTRs or 5′ UTRs (Fig. 5B). Further, when we compared the DNA CpG methylation levels from control, TPA and TPA + CA group, the average percent methylation in promotor region was lowest while 3′UTR and 5′UTR showed relatively higher levels within all three groups (Fig. 5C). However, no significant difference was observed in average percent methylation amongst all three groups (Control, TPA and TPA + CA).

Fig. 5.

Fig. 5.

Open in a new tab

Methylseq analysis representing the alterations induced by TPA and CA in JB6 cells. A) Scattered distribution of annotated DMRs by its gene feature. B) DMRs distribution by number and region of CpGs. C) Average methylation levels of DMRs based on gene regions for samples in the control, TPA, and TPA + CA groups.

3.6. Effect on overall DMRs by TPA and its modulation by CA treatment

The changes in the methylation profiles between TPA treated vs control group and TPA + CA group with only TPA group were studied to understand the effect of TPA and CA on JB6P + cells. The methylation difference of threshold of more than 0.1 was used to identify the DMRs. The differences in the methylation were investigated between the three treatment groups viz. control, TPA treated and TPA + CA treated groups in two comparisons, positive and negative differences. As shown in Fig. 6A, 167 genes were significantly hypermethylated while 187 genes were hypomethylated by TPA as compared to control group. While 133 genes are hypermethylated and 155 genes were hypomethylated by CA as compared to TPA only treated group. As shown in Fig. 6A DMRs in the promoter region (≤1 Kb) were significantly hypermethylated by TPA (28/167) but reversed and hypomethylated by CA treatment (28/155). While DMRs in the promoter region (≤1 Kb) were hypomethylated by TPA (15/187) and reversed and hypermethylated by CA treatment (15/118). Interestingly, genes like Smad-3 (signal transducers for transforming growth factor beta (TGF-B) superfamily), Tasp1 gene (endopeptidase which needs for the maintenance of HOX genes), Uri1 (prefoldin like chaperone protein), Nsg2 (Neuronal vesicle trafficking-associated protein 2) shown to be hypermethylated by TPA treatment while hypomethylated by CA (Fig. 6B). While genes like Madd (MAP kinase activating death domain), Dusp22 (Dual specificity protein phosphatase 22), Rassf (tumor suppressor gene family, Ras-association domain family) genes were found to be hypomethylated by TPA but hypermethylated by CA treatment (Fig. 6B).

Fig. 6.

Fig. 6.

Open in a new tab

Overview of DNA methylation modification by TPA treatment and CA treatment. A) Venn diagrams of comparing the hypermethylated and hypomethylated genes in both the control versus TPA group and the TPA only vs. TPA + CA groups. Genes with q < 0.05 and log2 (fold change) > 1 or < −1 were considered for the given analysis. B) Heat map showing the genes with Differentially methylated regions (DMRs) in both control vs TPA and TPA vs TPA + CA group comparisons.

3.7. Correlating the gene expression transcriptome and DNA methylome

It is widely known that promotor DNA methylation of DMRs is related to the lowered transcription and hence expression of the associated genes. To understand the association between the transcription and methylation, we tried to integrated these profiles with the threshold of 0.1 for DNA methylation with two-fold changes for gene expression 15 DEGs/DMRs were identified in the TPA group versus control group comparison (Fig. 7A) while with the same threshold, 15 genes were identified in TPA vs TPA + CA group (Fig. 7B). Each dot in figure represents a DMR/DEG with corresponding features of the gene. Integrative analysis of RNA-seq and DNA-seq data yield a subset of genes associated with TPA and CA treatment, and the gene expression changes driven by promoter CpG status. The genes involve like CDK1(Cyclin-dependent kinases 1), GDF11(Growth Differentiation Factor 11), RASSF2 (Ras association domain family member 2), Dock8 (Dedicator of cytokinesis protein 8) were identified.

Fig. 7.

Fig. 7.

Open in a new tab

Integration between transcriptome and DNA methylation A) Scatter plot with DMRs/genes in the control versus TPA group with cutoff of 0.1 for DNA methylation and a 2-fold change threshold for RNA expression B) Scatter plot demonstrating DMRs in the TPA versus TPA + CA group comparison based on a cutoff of 0.1 for DNA methylation and a 1-fold change threshold for RNA expression. The DMRs locations are indicated by the colors.

3.8. Validating the differentially regulated genes by qPCR

CDK-1, GDF11, DOCK-8, and RASSF-1 were selected for qPCR validation. The oligo primers are summarized in Supplementary Table S2. The relative mRNA expression levels of the genes are presented in Fig. 8 qPCR showed a similar pattern consistent with the RNA-seq results, suggesting that the NGS approach used in this study can appropriately represent a transcriptional expression.

Fig. 8.

Fig. 8.

Open in a new tab

Differentially regulated genes from the RNAseq data were validated by qPCR. Validation of relative RNA expression in TPA and TPA + CA groups.

4. Discussion

Carcinogenesis is multifactorial and long process consisting of initiation, promotion, and progression stages [13]. Skin cancer progresses through various stages including basal cell carcinoma (BCC), squamous cell carcinoma (SCC), and melanoma [33]. Oxidative stress along with chronic inflammation is known to drive the process of carcinogenesis. The conversion of initiated cells into tumorigenic cells by various tumor promoters, potentially by promoting oxidative/in-flammatory responses is reported [34–36]. Genetic mutations during the progression of skin carcinogenesis is well documented [37], however the epigenetic changes along with gene expression studies induced by TPA remains elusive. Many natural phytochemicals, such as curcumin [38], sulforaphane [39] and green tea [40] have been shown to have cancer chemopreventive activity against skin cancer. The triterpenoid, corosolic acid, found in many medicinal herbs and its anti-diabetic activity is found in many food supplementary products [41]. In the given study, we demonstrated that TPA induced transcriptomic and methylation changes using JB6 cells which was reversed by CA treatment. Moreover, various novel genes involved in the process of skin carcinogenesis were reported for the first time which were demonstrated as differentially methylated, including GDF11, Dock8, Madd, Rassf. Previous study showed CA perturbed the cell-cell interactions between macrophages and ovarian epithelial cells and shown to inhibit the macrophage-induced activation of ovarian cancer cells [20]. Corosolic acid inhibited the CD163 expression, known phenotype markers of M2 macrophages, and shown to suppress the secretion of IL-10 [19]. Another study had shown corosolic acid dose-dependently inhibited the viability of HCT116 cells. CA overexpressed the pro-apoptotic proteins, such as Bax, Fas and FasL while decreased the expression of anti-apoptotic proteins, such as Bcl-2 and survivin [16]. Our study has also demonstrated the dose dependent decreased viability of JB6 cells during 1,3 and 5 days by MTS assay. Pretreatment with CA had shown to decrease the phosphorylation levels of focal adhesion kinase (FAK) and ERK1/2, suggesting that CA has anti-angiogenic activity that which suppress FAK signaling induced by angiopoietin-1 [42]. Previous studies using CA has been demonstrated STAT3 inhibitor, which was able to increase the sensitivity to chemotherapeutic drugs in epithelial ovarian cancer cells [20]. Our study has also shown the various angiogenesis proteins like Prolactin-2C2 and Sprr2h (small proline-rich protein 2) known to be regulated by IL6/STAT3 signaling are also get deregulated by TPA treatment which was reversed by CA treatment. Importantly, various pathways including rho family GTPase, PEDF signaling pathway, TNFR2 signaling pathway which was upregulated by TPA treatment was further downregulated by CA treatment.

Epigenetic aberrations, including differences in DNA methylation patterns or changes in histone modifications, are major key drivers of skin carcinogenesis [43]. Our previous report on DNA methylation analysis revealed that UA demethylated the first 15 CpG sites of the Nrf2 promoter region, which correlated with the expression of Nrf2 in JB6 P+ cells [44]. Structural similarity between urosolic acid and corosolic acid prompted to analyze the epigenetic modifications using JB6 cells. Interestingly, in the present study the genes like Smad-3 (signal transducers for transforming growth factor beta (TGF-B) superfamily), Tasp1 gene (endopeptidase which needs for the maintenance of HOX genes), Uri1 (prefoldin like chaperone protein), Nsg2 (Neuronal vesicle trafficking-associated protein 2) shown to be hypermethylated by TPA treatment while hypomethylated by CA. Previous report also suggests Smad 3 transmits TGF-β signals from the receptor to the COL1A2 promoter in human skin fibroblasts, and is important in stimulation of COL1A2 promoter activity elicited by TGF-β [45]. Our methylseq analysis has also emerged the novel epigenetic targets in the early stages of skin carcinogenesis like Madd (MAP kinase activating death domain), Dusp22 (Dual specificity protein phosphatase22), Rassf (tumor suppressor gene family, Ras-association domain family) genes were found to be hypomethylated by TPA but hypermethylated by CA treatment. Skin carcinogenesis is a complex process regulated by an multiple interlinked network of genes and pathways [46]. We also analyzed crucial subset of genes which are associated with the chemopreventive effects of CA in TPA-induced JB6P + cells were elucidated by exploring the correlations between gene expression and DNA methylation associated with the process. These novel insights will facilitate the discovery of important therapeutic targets in the relevant mechanisms and suggesting potential biomarkers for the prevention/treatment of skin carcinogenesis. However, the further rigorous validation and comprehensive mechanistic studies using various model systems are essential for the robust conclusion about these genes which are potentially involved in the process of skin carcinogenesis and chemoprevention.

In summary, given study demonstrates the chemopreventive effect of CA against TPA induced early transformation in mouse epidermal JB6P + cells. Using DNA methylseq and RNA-seq, we profiled the quantitative global methylome and transcriptome with and without CA treatment. A battery of potential transcriptomic and epigenomic biomarkers has been uncovered to be modulated in the progression of the skin carcinogenesis process. These findings provide important insights into the way epigenetic modifications affect the progression of skin carcinogenesis and into the preventive effects of CA.

Supplementary Material

1
2

Acknowledgments

This study was supported by grant R01 CA200129, from the National Cancer Institute awarded to Dr. Ah-Ng Tony Kong. The authors thank all the members of Dr. Kong’s laboratory for their invaluable support and technical assistance.

Abbreviations:

CA

Corosolic acid

DEGs

differentiated expressed genes

IPA

Ingenuity Pathway analysis

TSS

transcription initiation site

NGS

next generation sequencing

NMSC

Non-melanoma skin cancer

TPA

12-O-tetradecanoyl-phorbol-13-acetate

NQO1

NAD (P) H Quinone Oxidoreductase-1

Footnotes

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.cbi.2020.109025.

References

  • [1].Diepgen TL, Mahler V, The epidemiology of skin cancer, Br. J. Dermatol 146 (61) (2002) 1–6. [DOI] [PubMed] [Google Scholar]
  • [2].Greenberg ES, Chong KK, Huynh KT, Tanaka R, Hoon DS, Epigenetic biomarkers in skin cancer, Canc. Lett 342 (2014) 170–177. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [3].Yang Y, Wu R, Sargsyan D, Yin R, Kuo HC, Yang I, Wang L, Cheng D, Wang C,Li S, Hudlikar R, Lu Y, Kong AN, UVB drives different stages of epigenome alterations during progression of skin cancer, Canc. Lett 449 (2019) 20–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [4].Schinke C, Mo Y, Yu Y, Amiri K, Sosman J, Greally J, Verma A, Aberrant DNA methylation in malignant melanoma, Melanoma Res. 20 (2010) 253–265. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [5].Hudlikar RR, Pai V, Kumar R, Thorat RA, Kannan S, Ingle AD, Maru GB, Mahimkar MB, Dose-related modulatory effects of polymeric black tea polyphenols (PBPs) on initiation and promotion events in B(a)P and NNK-induced lung carcinogenesis, Nutr. Canc 71 (2019) 508–523. [DOI] [PubMed] [Google Scholar]
  • [6].Chinembiri TN, du Plessis LH, Gerber M, Hamman JH, du Plessis J, Review of natural compounds for potential skin cancer treatment, Molecules 19 (2014) 11679–11721. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [7].Hsu TC, Nair R, Tulsian P, Camalier CE, Hegamyer GA, Young MR,Colburn NH, Transformation nonresponsive cells owe their resistance to lack of p65/nuclear factor-kappaB activation, Canc. Res 61 (2001) 4160–4168. [PubMed] [Google Scholar]
  • [8].Colburn NH, Gindhart TD, Hegamyer GA, Blumberg PM, Delclos KB,Magun BE, Lockyer J, Phorbol diester and epidermal growth factor receptors in 12-O-tetradecanoylphorbol-13-acetate-resistant and -sensitive mouse epidermal cells, Canc. Res 42 (1982) 3093–3097. [PubMed] [Google Scholar]
  • [9].Colburn NH, Bruegge WF, Bates JR, Gray RH, Rossen JD, Kelsey WH,Shimada T, Correlation of anchorage-independent growth with tumorigenicity of chemically transformed mouse epidermal cells, Canc. Res 38 (1978) 624–634. [PubMed] [Google Scholar]
  • [10].Bernstein LR, Colburn NH, AP1/jun function is differentially induced in promotion-sensitive and resistant JB6 cells, Science 244 (1989) 566–569. [DOI] [PubMed] [Google Scholar]
  • [11].Sandoval J, Esteller M, Cancer epigenomics: beyond genomics, Curr. Opin. Genet. Dev 22 (2012) 50–55. [DOI] [PubMed] [Google Scholar]
  • [12].Allis CD, Jenuwein T, The molecular hallmarks of epigenetic control, Nat. Rev. Genet 17 (2016) 487–500. [DOI] [PubMed] [Google Scholar]
  • [13].Maru GB, Hudlikar RR, Kumar G, Gandhi K, Mahimkar MB, Understanding the molecular mechanisms of cancer prevention by dietary phytochemicals: from experimental models to clinical trials, World J. Biol. Chem 7 (2016) 88–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [14].Lee JH, Khor TO, Shu L, Su ZY, Fuentes F, Kong AN, Dietary phytochemicals and cancer prevention: Nrf2 signaling, epigenetics, and cell death mechanisms in blocking cancer initiation and progression, Pharmacol. Ther 137 (2013) 153–171. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [15].Shankar E, Kanwal R, Candamo M, Gupta S, Dietary phytochemicals as epigenetic modifiers in cancer: promise and challenges, Semin. Canc. Biol 40–41 (2016) 82–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [16].Sung B, Kang YJ, Kim DH, Hwang SY, Lee Y, Kim M, Yoon JH, Kim CM, Chung HY, Kim ND, Corosolic acid induces apoptotic cell death in HCT116 human colon cancer cells through a caspase-dependent pathway, Int. J. Mol. Med 33 (2014) 943–949. [DOI] [PubMed] [Google Scholar]
  • [17].Kim JH, Kim YH, Song GY, Kim DE, Jeong YJ, Liu KH, Chung YH, Oh S, Ursolic acid and its natural derivative corosolic acid suppress the proliferation of APC-mutated colon cancer cells through promotion of beta-catenin degradation, Food Chem. Toxicol 67 (2014) 87–95. [DOI] [PubMed] [Google Scholar]
  • [18].Xu Y, Ge R, Du J, Xin H, Yi T, Sheng J, Wang Y, Ling C, Corosolic acid induces apoptosis through mitochondrial pathway and caspase activation in human cervix adenocarcinoma HeLa cells, Canc. Lett 284 (2009) 229–237. [DOI] [PubMed] [Google Scholar]
  • [19].Fujiwara Y, Komohara Y, Ikeda T, Takeya M, Corosolic acid inhibits glioblastoma cell proliferation by suppressing the activation of signal transducer and activator of transcription-3 and nuclear factor-kappa B in tumor cells and tumor-associated macrophages, Canc. Sci 102 (2011) 206–211. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [20].Fujiwara Y, Takaishi K, Nakao J, Ikeda T, Katabuchi H, Takeya M, Komohara Y, Corosolic acid enhances the antitumor effects of chemotherapy on epithelial ovarian cancer by inhibiting signal transducer and activator of transcription 3 signaling, Oncol. Lett 6 (2013) 1619–1623. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [21].Lee HS, Park JB, Lee MS, Cha EY, Kim JY, Sul JY, Corosolic acid enhances 5-fluorouracil-induced apoptosis against SNU-620 human gastric carcinoma cells by inhibition of mammalian target of rapamycin, Mol. Med. Rep 12 (2015) 4782–4788. [DOI] [PubMed] [Google Scholar]
  • [22].Ku CY, Wang YR, Lin HY, Lu SC, Lin JY, Corosolic acid inhibits hepatocellular carcinoma cell migration by targeting the VEGFR2/src/FAK pathway, PloS One 10 (2015) e0126725. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [23].Yang J, Wu R, Li W, Gao L, Yang Y, Li P, Kong AN, The triterpenoid corosolic acid blocks transformation and epigenetically reactivates Nrf2 in TRAMP-C1 prostate cells, Mol. Carcinog 57 (2018) 512–521. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [24].Yang Y, Yang I, Cao M, Su ZY, Wu R, Guo Y, Fang M, Kong AN, Fucoxanthin elicits epigenetic modifications, Nrf2 activation and blocking transformation in mouse skin JB6 P+ cells, AAPS J. 20 (2018) 32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [25].Martin M, Cutadapt removes adapter sequences from high throughput sequencing reads, EMBnet 17 (2011) 10–12. [Google Scholar]
  • [26].Kim D, Langmead B, Salzberg SL, HISAT: a fast spliced aligner with low memory requirements, Nat. Methods 12 (2015) 357–360. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [27].Liao Y, Smyth GK, Shi W, featureCounts: an efficient general purpose program for assigning sequence reads to genomic features, Bioinformatics 30 (2014) 923–930. [DOI] [PubMed] [Google Scholar]
  • [28].Wang L, Feng Z, Wang X, Wang X, Zhang X, DEGseq: an R package for identifying differentially expressed genes from RNA-seq data, Bioinformatics 26 (2010) 136–138. [DOI] [PubMed] [Google Scholar]
  • [29].Guo Y, Wu R, Gaspar JM, Sargsyan D, Su ZY, Zhang C, Gao L, Cheng D, Li W,Wang C, Yin R, Fang M, Verzi MP, Hart RP, Kong AN, DNA methylome and transcriptome alterations and cancer prevention by curcumin in colitis-accelerated colon cancer in mice, Carcinogenesis 39 (2018) 669–680. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [30].Krueger F, Andrews SR, Bismark: a flexible aligner and methylation caller for Bisulfite-Seq applications, Bioinformatics 27 (2011) 1571–1572. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [31].Gaspar JM, Hart RP, DMRfinder: efficiently identifying differentially methylated regions from MethylC-seq data, BMC Bioinf. 18 (2017) 528. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [32].Yu G, Wang LG, He QY, ChIPseeker: an R/Bioconductor package for ChIP peak annotation, comparison and visualization, Bioinformatics 31 (2015) 2382–2383. [DOI] [PubMed] [Google Scholar]
  • [33].Katalinic A, Kunze U, Schafer T, Epidemiology of cutaneous melanoma and non-melanoma skin cancer in Schleswig-Holstein, Germany: incidence, clinical sub-types, tumour stages and localization (epidemiology of skin cancer), Br. J. Dermatol 149 (2003) 1200–1206. [DOI] [PubMed] [Google Scholar]
  • [34].Ito N, Hirose M, Antioxidants–carcinogenic and chemopreventive properties, Adv. Canc. Res 53 (1989) 247–302. [DOI] [PubMed] [Google Scholar]
  • [35].Pillai CK, Pillai KS, Antioxidants in health, Indian J. Physiol. Pharmacol 46 (2002) 1–5. [PubMed] [Google Scholar]
  • [36].Hudlikar RR, Venkadakrishnan VB, Kumar R, Thorat RA, Kannan S, Ingle AD,Desai S, Maru GB, Mahimkar MB, Polymeric black tea polyphenols (PBPs) inhibit benzo(a)pyrene and 4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone-induced lung carcinogenesis potentially through down-regulation of p38 and Akt phosphorylation in A/J mice, Mol. Carcinog 56 (2017) 625–640. [DOI] [PubMed] [Google Scholar]
  • [37].de Gruijl FR, van Kranen HJ, Mullenders LH, UV-induced DNA damage, repair, mutations and oncogenic pathways in skin cancer, J. Photochem. Photobiol., B 63 (2001) 19–27. [DOI] [PubMed] [Google Scholar]
  • [38].Limtrakul P, Lipigorngoson S, Namwong O, Apisariyakul A, Dunn FW, Inhibitory effect of dietary curcumin on skin carcinogenesis in mice, Canc. Lett 116 (1997) 197–203. [DOI] [PubMed] [Google Scholar]
  • [39].Xu C, Huang MT, Shen G, Yuan X, Lin W, Khor TO, Conney AH, Kong AN, Inhibition of 7,12-dimethylbenz(a)anthracene-induced skin tumorigenesis in C57BL/6 mice by sulforaphane is mediated by nuclear factor E2-related factor 2, Canc. Res 66 (2006) 8293–8296. [DOI] [PubMed] [Google Scholar]
  • [40].Wang ZY, Huang MT, Lou YR, Xie JG, Reuhl KR, Newmark HL, Ho CT, Yang CS, Conney AH, Inhibitory effects of black tea, green tea, decaffeinated black tea, and decaffeinated green tea on ultraviolet B light-induced skin carcinogenesis in 7,12-dimethylbenz[a]anthracene-initiated SKH-1 mice, Canc. Res 54 (1994) 3428–3435. [PubMed] [Google Scholar]
  • [41].Stohs SJ, Miller H, Kaats GR, A review of the efficacy and safety of banaba (Lagerstroemia speciosa L.) and corosolic acid, Phytother Res. 26 (2012) 317–324. [DOI] [PubMed] [Google Scholar]
  • [42].Yoo KH, Park JH, Lee DY, Hwang-Bo J, Baek NI, Chung IS, Corosolic acid exhibits anti-angiogenic and anti-lymphangiogenic effects on in vitro endothelial cells and on an in vivo CT-26 colon carcinoma animal model, Phytother Res. 29 (2015) 714–723. [DOI] [PubMed] [Google Scholar]
  • [43].Eisenstein A, Gonzalez EC, Raghunathan R, Xu X, Wu M, McLean EO,McGee J, Ryu B, Alani RM, Emerging biomarkers in cutaneous melanoma, Mol. Diagn. Ther 22 (2018) 203–218. [DOI] [PubMed] [Google Scholar]
  • [44].Kim H, Ramirez CN, Su ZY, Kong AN, Epigenetic modifications of triterpenoid ursolic acid in activating Nrf2 and blocking cellular transformation of mouse epidermal cells, J. Nutr. Biochem 33 (2016) 54–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [45].Chen SJ, Yuan W, Mori Y, Levenson A, Trojanowska M, Varga J, Stimulation of type I collagen transcription in human skin fibroblasts by TGF-beta: involvement of Smad 3, J. Invest. Dermatol 112 (1999) 49–57. [DOI] [PubMed] [Google Scholar]
  • [46].Roop DR, Krieg TM, Mehrel T, Cheng CK, Yuspa SH, Transcriptional control of high molecular weight keratin gene expression in multistage mouse skin carcinogenesis, Canc. Res 48 (1988) 3245–3252. [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1572695-supplement-1.docx (13.4KB, docx)
2
NIHMS1572695-supplement-2.docx (10.9KB, docx)

RESOURCES